Re: [sca-assembly] Exit Criteria for the SCA Specifications.

[Anish Karmarkar <Anish.Karmarkar@oracle.com>]

I'm confused about how does this differ from your proposal #1 [1], as 
far as the optional normative parts are concerned. All this proposal 
does is amend you proposal #1 to address some of the points raised on 
the call. Specifically,
a) address Eric's point about whether each implementation has to 
implement *everything* normative, or can there be 2 implementations of 
each "feature". I.e., can N implementations satisfy the criteria even 
though none of the N impls implement everything, but between them they 
have 2 implementations of every normative item.
b) address Danny's issue of how do we verify that someone implements a 
particular feature/conformance item.

Certainly, this proposal (as well as your, as you point out in your 
email) have the issues of --
1) this requires that we have implementations that demonstrate 
implementability of the optional bits
2) requires two implementations to run the test suite for the specs to 
move to the next stage (exit criteria not conformance criteria).

Both of these are good discussion points.

To you central point in the email wrt the optional bits:
I think the TC needs to decide whether we are going to use W3C's model 
(that has been proposed  by Jeff) or not. The W3C model is to extract 
"atomic features" from the spec (regardless of whether it is optional or 
mandatory) and set acceptance level. For example, 5 impls for each 
mandatory feature and 2 impls for each optional feature. Any feature 
that does not meet the requisite acceptance level is marked as 'feature 
at risk'. By some reasonable deadline, if no one steps forward to put in 
the effort to reach the acceptance level then the feature gets deleted 
from the spec.

We don't necessarily have to follow this model. But the charter does 
require that there be two implementations of optional and mandatory 
normative features (they don't have to pass the test suite). I think we 
have to either change the charter if we dont' like this OR figure out 
the mechanism the TC is going to use to accept that two implementations 
have indeed implemented the normative bits (optional and mandatory). WRT 
the mechanism, we can decide to accept statements from implementers 
stating which features they implement (kind of like the OASIS 'statement 
of use'), ask for results of running our test suite, or something else.

-Anish
--


[1] http://lists.oasis-open.org/archives/sca-assembly/201103/msg00029.html

On 3/9/2011 3:10 AM, Mike Edwards wrote:
>
> Folks,
>
> My view of this proposal is that it makes optional normative parts of
> any specifications pretty well untenable.
> Perhaps this is the intention.
>
> If this was what was intended from the start, then I am sure that the
> approach to optional sections of any of the
> SCA specifications in any of the TCs would have been very different.
>
> The implication is that there would be an arbitrarily high bar set to
> achieve the exit criteria for the SCA specifications.
> I see no point to this - it is a recipe for never reaching standardization.
>
> I see two alternative paths:
>
> o Not accept this approach to the Exit Criteria
>
> o Start now removing all the optional normative parts of the SCA
> specifications (either remove them entirely or make
> them non-normative - which of these to adopt will vary depending on the
> specification).
>
>
> I remind everyone that the impact of this decision will be greater in
> some of the other TCs - some specifications such
> as the Web Services one - have significant optional sections. It will
> also have an impact on the development of the
> 1.2 version of the Assembly specification.
>
>
> Yours, Mike
> ------------------------------------------------------------------------
> Dr Mike Edwards 	Mail Point 137, Hursley Park 	
> STSM 	Winchester, Hants SO21 2JN
> SCA & Services Standards 	United Kingdom
> Co-Chair OASIS SCA Assembly TC 		
> IBM Software Group 		
> Phone: 	+44-1962 818014 		
> Mobile: 	+44-7802-467431 (274097) 		
> e-mail: 	mike_edwards@uk.ibm.com 		
>
>
>
>
>
> From: 	Anish Karmarkar <Anish.Karmarkar@oracle.com>
> To: 	Danny van der Rijn <dannyv@tibco.com>
> Cc: 	OASIS Assembly <sca-assembly@lists.oasis-open.org>
> Date: 	08/03/2011 17:24
> Subject: 	[sca-assembly] Amendment to the amendment of today's motion
>
>
> ------------------------------------------------------------------------
>
>
>
> Danny,
>
> Here is the text from the chat:
>
> "o the 2 independent runtimes shall pass the Test Suite for SCA Assembly
> as described in the document 'TestCases for the SCA Assembly Model
> Version 1.1 Specification'. The TC shall use the results (expected
> output) of the relevant tests, submitted by the implementer (in any
> form), to verify that the runtime passes the tests."
>
>
> This + changes to talk about pair-wise (instead of all) implementation
> of normative parts would change Mike's proposal 1 to:
>
> "The Concrete Exit Criteria for the SCA Assembly V1.1 specification are
> that:
>
> o there shall be at least 2 independent SCA runtimes that are compliant
> with each normative portion of the specification as described in Section
> 12.2 of the SCA Assembly V1.1 specification
>
> o the 2 independent runtimes shall pass the Test Suite for SCA Assembly
> as described in the document 'TestCases for the SCA Assembly Model
> Version 1.1 Specification'. The TC shall use the results (expected
> output) of the relevant tests, submitted by the implementer (in any
> form), to verify that the runtime passes the tests.
>
> -Anish
> --
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail. Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
>
>
>
>
> ------------------------------------------------------------------------
>
> /
> /
>
> /Unless stated otherwise above:
> IBM United Kingdom Limited - Registered in England and Wales with number
> 741598.
> Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU/
>
>
>
>
>
>